|
Veterans Alcoholism Screening Test (VAST)
Please note that this section is an archive (last updated in June 2006). [disclaimer]
Sections:
Overview |
Instrument Reviews |
Construct Overviews |
Book Compendium Reviews |
Internet Site Reviews
|
|
|
|
Practical Information
|
Instrument Name:
|
Veterans Alcoholism Screening Test (VAST)
|
Instrument Description:
|
The VAST was developed at the Veterans Administration Medical Center in Fayetteville, North Carolina and is used: (1) to measure alcoholism in adults, (2) as a screening device for alcoholism, or (3) for research purposes. It is a refinement of the Michigan Alcoholism Screening Test (MAST). The MAST does not distinguish between past and current problems with alcoholism; therefore, there was an unmet need for a version that did. The VAST was designed to distinguish between past and current alcoholism by retaining the original 24 MAST items, and adding 3 questions that identify specific time periods for responses indicating alcoholism. These three questions follow each of the 24 original MAST items. The 3 additional questions are: 1) within the last year? (current), 2) >1 year ago but <=5 years ago?, and 3) >5 years ago?. (Ref: 1-2, 4) The additional questions allow scores for the three specified time periods to be obtained, as well as the original MAST score. The VAST is especially well suited to assess alcoholism among older adults because it distinguishes among recent, near past, and far distant alcohol use. (Ref: 2)
|
Price:
|
Free (available in the developmental article; Ref: 1)
|
Administration Time:
|
5-10 minutes; the VAST takes only 3 minutes longer to administer than the MAST. (Ref: 1)
|
Publication Year:
|
1982
|
Item Readability:
|
No information found. Items appear to use language understandable to the normal adult population and seem self-explanatory.
|
Scale Format:
|
Dichotomous yes/no, and a few open-ended response options.
|
Administration Technique:
|
Self-administered paper and pencil or interview. No training is required for administration.
|
Scoring and Interpretation:
|
The weights and scoring for each VAST item are the same as for each MAST item. Four (4) scores may be obtained: 1) MAST score, 2) VAST-C (current alcoholism), 3) VAST 1-5 (alcoholism in the past 5 years), and 4) VAST>5 (alcoholism greater than 5 years ago). (Ref: 1) Items are weighted heavier for more indicative symptoms, and points are assigned to the time period reported. A score of 6 or more (in a particular time period) on each of these measures suggests alcoholism. (Ref: 1, 2, 4)
|
Forms:
|
Original MAST
|
|
|
Research Contacts
|
Instrument Developers:
|
Kathryn Magruder-Habib, Katherine Harris, and Glynn Fraker. Copyright is held by the Journal of Studies on Alcohol Editorial Office, Rutgers University Center of Alcohol Studies, New Brunswick, NJ.
|
Instrument Development Location:
|
Health Services Research Service
Veterans Affairs Medical Center
2300 Ramsey Street
Fayetteville , NC 28301
|
Instrument Developer Email:
|
No information found.
|
Instrument Developer Website:
|
No information found.
|
|
|
Annotated Bibliography
|
1. Magruder-Habib K, Harris KE, Fraker GG. Validation of the Veterans Alcoholism Screening Test. J Stud Alcohol. 1982 Sep;43(9):910-26. [PMID: 7166959]
 |
Purpose: To determine the concurrent validity of the VAST. |
|
Sample: 118 pairs of ambulatory care patients and their relatives, attending the VA in North Carolina. Average age was 59 years. Ninety-two (78.0%) of the 118 relatives interviewed were wives, 10 (8.5%) were children, 4 (3.4%) were parents, 3 (2.5%) were siblings, 3 (2.5%) were in-laws, and 6 (5.1%) were in a category of “other”. |
|
Methods: Subjects and relatives were interviewed independently, and subjects were given the VAST while relatives were given the VAST reworded to elicit responses about the subject. Scores and reports were compared. |
|
Implications: Between pair correlations were .767 (VAST-C), .738 (VAST 1-5), and .396 (VAST >5). These values compare to a between pair score of .492 for the MAST. There was higher inter-pair agreement in VAST scores were better than in MAST scores in categorical assignment (using a cut score of > or = to cut score of 6—no justification for this particular cut score is given).Implications: There was greater agreement between patients’ scores and relatives’ reports on the VAST-C and VAST 1-5, but not on the VAST >5. The response format of the distinguishes between past and current alcoholism while the response format of the MAST does not. |
2. Magruder-Habib K, Saltz CC, Barron PM. Age-related patterns of alcoholism among veterans in ambulatory care. Hosp Commun Psychiatry 1986;37(12)1251-55. [PMID: 3492420]
|
Purpose: To examine age-related prevalence of alcoholism in outpatient veterans. |
|
Sample: 432 patients in two urban VA medical centers, 90 (20.8%) of whom were missing data on age. Data reported are for the 342 remaining patients. . Most (333, 97.4%) of the patients were male. Based on a denominator of 324 patients for whom marital status was available, 158 (48.8%) lived with their spouses. Ethnicity data were not provided. |
|
Methods: Subjects were randomly sampled during a one-week period, then given the VAST and a questionnaire eliciting demographic data and alcohol treatment information for the previous 30 days for a study estimating alcoholism prevalence among the VA outpatient population. Alcoholism was operationalized as a score of 6 or higher on the VAST. |
|
Implications: Based on the VAST cut off score of 6 or higher, half of the 65 years and older alcoholics were long standing alcoholic. The younger age groups (under 35, 35 to 44, and 45 to 54) were two and a half to three times more likely to be categorized as alcoholic. Just over 10% of the study sample who were 65 or over were categorized by their VAST score as alcoholic. |
3. Adams SL, Waskel SA. Late onset alcoholism: stress or structure. J of Psychology 1993;127(3):329-34. [PMID: 8377146]
|
Purpose: To ascertain whether stress related to a loss of social psychological structures was greater for later onset alcoholics than for other individuals. |
|
Sample: 60 consecutively admitted males from six Midwest residential alcohol treatment centers, mean age=63.6 years. These men all had a primary diagnosis of alcoholism, and reported lower incomes than the general older adult population (less than $15,000). |
|
Methods: Subjects completed the VAST and the Drinking Practices Survey, and were then identified as early onset (before age 40 years, n=33), quasi-late onset (between the ages of 40 and 60 years, n=20), or later onset (after age 60 years, n=7)) alcoholics. Due to small sample sizes in the latter two groups, the data were later structured into the early onset group (n=33) and the late onset group (n=27). |
|
Implications: Chi-square cross-tabulations indicated that a higher number of early onset alcoholics were married, compared to later onset alcoholics. Of the social-psychological structures, only marital status was significantly related to alcoholism. However, the sample, which is based on consecutive admissions, may not be representative of the study population. |
4. Magruder-Habib K, Stevens HA, Alling WC. Relative performance of the MAST, VAST, and CAGE versus DSM-III-R criteria for alcohol dependence. J Clin Epidemiol 1993;46(5):435-41. [PMID: 8501469]
|
Purpose: To determine the alcohol dependence screening effectiveness of the MAST, VAST, and CAGE in a general medical clinic at the VA Medical Center in Durham, N.C. in comparison to the DSM-III-R criteria for alcohol dependence as measured in the Composite International Diagnostic Interview (CIDI). |
|
Sample: Of the 1230 eligible patients, 915 (74.3%) signed the study’s consent form. Of the 915 eligible patients, 35 were identified as positive for alcoholism by the VAST (VAST-C $5) or by chart review. Of the 35 positive patients, complete information (including results from the CIDI and all screening tests) was available for 25 patients. Of the 880 negatives, completed information was available for 87 patients. Therefore, the sample contained 112 patients (87+25). The 112 patients had a mean age=54.4 years. Seventy-one (63.4%) of the patients were white, 74 (66.6%) were unemployed, and 75 (67.3%) were married. |
|
Methods: Trained research assistants administered the CAGE, CIDI-alcohol section, VAST and MAST. The VAST and CAGE were given first and second, and those patients positive for present alcoholism on the VAST or who had alcoholism for the previous 6 months noted in their medical chart, were then given the CIDI. Sensitivity, specificity, predictive values, accuracy, and Receiver Operating Characteristics were compared for all tests. |
|
Implications: All tests performed at acceptable levels, but the MAST and VAST had higher performance characteristics (higher sensitivity and specificity). However, the CAGE is the briefest and easiest to administer. |
|
|
top
|
Factors and Norms
|
Factor Analysis Work:
|
No information found.
|
Normative Information Availability:
|
No information found.
|
|
|
Reliability Evidence
|
Test-retest:
|
No information found.
|
Inter-rater:
|
No information found.
|
Internal Consistency:
|
No information found.
|
Alternate Forms:
|
No information found.
|
|
|
Validity Evidence
|
Construct/ Convergent/ Discriminant:
|
No information found.
|
Criterion-related/ Concurrent/ Predictive:
|
Concurrent validity was studied by comparing the relationship between scores provided by patients and scores provided by the patients’ relatives. Correlations were (see Instrument Description for an explanation of acronyms):
| Score
|
r
|
| VAST-C
|
0.767
|
| VAST 1-5
|
0.738
|
| MAST
|
0.492
|
| VAST > 5
|
0.396
|
When MAST and VAST-C scores were classified as indicative of alcoholism (>=6) or not (<6), the percentage of agreement was:: VAST-C=88.9%, VAST 1-5=83.8%, MAST=79.5%, VAST>5=72.6%. (Ref: 1)
When ROC analysis was performed, the VAST had an area under the curve (AUC) of 91.6% (±2.6%) and the VAST-C showed an AUC of 93.3% (± 3.4%). (Ref: 4)
|
Content:
|
No information found.
|
Responsiveness Evidence:
|
At a cut point of >=5, the VAST has sensitivity=95.1% and specificity=80.3%. At the cut point of >=5, the VAST-C shows a sensitivity of 83.3% and a specificity of 89.0%. (Ref: 4)
|
Scale Application in VA Populations:
|
Yes. (Ref: 1-2, 4)
|
Scale Application in non-VA Populations:
|
Yes. (Ref: 3)
|
|
|
Comments
|
The strengths of the VAST are it’s discrimination between present and past alcohol abuse and it’s high concordance with the DSM-III-R. Though the original article (ref #1) offered no explanation for its recommended cut point of 6, the Magruder-Habib study calculated the sensitivity and specificity of cut points of1-8. The purpose of screening instruments is to inexpensively identify persons who may have a condition. Generally, persons who are positive on the basis of a screen receive follow-up testing. The VAST has been used for a different purpose—to classify persons as alcoholic or not. When it is used in this way, justification should be given for the choice of cut score. It is important to be precise in describing classifications based on the VAST or any of the other alcohol screening measures (e.g., “Based on a VAST cut score of 5, 20% of the sample population were identified as alcoholic.”)
|
Updates
No information found.
|
|
|
|
